Effects of Management on Lichen Species Richness, Ecological ...

RESEARCH ARTICLE

Effects of Management on Lichen SpeciesRichness, Ecological Traits and CommunityStructure in the Rodnei Mountains NationalPark (Romania)Ioana Violeta Ardelean1,2*, Christine Keller1, Christoph Scheidegger1

1 Swiss Federal Research Institute WSL, 8903 Birmensdorf, Switzerland, 2 National Institute of Researchand Development for Biological Sciences, Cluj–Napoca, Romania

* [email protected]

AbstractLichens are valuable bio-indicators for evaluating the consequences of human activities

that are increasingly changing the earth’s ecosystems. Since a major objective of national

parks is the preservation of biodiversity, our aim is to analyse how natural resource man-

agement, the availability of lichen substrates and environmental parameters influence

lichen diversity in Rodnei Mountains National Park situated in the Eastern Carpathians.

Three main types of managed vegetation were investigated: the transhumance systems in

alpine meadows, timber exploitation in mixed and pure spruce forests, and the correspond-

ing conserved sites. The data were sampled following a replicated design. For the analysis,

we considered not only all lichen species, but also species groups from different substrates

such as soil, trees and deadwood. The lichen diversity was described according to species

richness, red-list status and substrate-specialist species richness. The variation in species

composition was related to the environmental variables. Habitat management was found to

negatively influence species richness and alter the lichen community composition, particu-

larly for threatened and substrate-specialist species. It reduced the mean level of threat-

ened species richness by 59%, when all lichen species were considered, and by 81%,

when only epiphytic lichens were considered. Management-induced disturbance signifi-

cantly decreased lichen species richness in forest landscapes with long stand continuity.

The diversity patterns of the lichens indicate a loss of species richness and change in spe-

cies composition in areas where natural resources are still exploited inside the borders of

the national park. It is thus imperative for protected areas, in particular old-growth forests

and alpine meadows, to receive more protection than they have received in the past to

ensure populations of the characteristic species remain viable in the future.

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 1 / 16

a11111

OPEN ACCESS

Citation: Ardelean IV, Keller C, Scheidegger C(2015) Effects of Management on Lichen SpeciesRichness, Ecological Traits and Community Structurein the Rodnei Mountains National Park (Romania).PLoS ONE 10(12): e0145808. doi:10.1371/journal.pone.0145808

Editor: Takeshi Miki, National Taiwan University,TAIWAN

Received: July 31, 2015

Accepted: December 9, 2015

Published: December 30, 2015

Copyright: © 2015 Ardelean et al. This is an openaccess article distributed under the terms of theCreative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in anymedium, provided the original author and source arecredited.

Data Availability Statement: All data are included inthe manuscript or cited accordingly.

Funding: This work was supported by SectoralOperational Programme for Human ResourcesDevelopment 2007-2013, co-financed by theEuropean Social Fund, under the project numberPOSDRU/107/1.5/S/76841 (http://granturi.ubbcluj.ro/76841/) and Federal Office for the Environment(FOEN) (http://www.bafu.admin.ch/index.html?lang=en). The funders had no role in study design, datacollection and analysis, decision to publish, orpreparation of the manuscript.

IntroductionThe loss of taxonomic, phylogenetic, genetic, and functional diversity in ecosystems worldwideis currently taking place very rapidly due to human activities [1]. To slow down the rate of loss,effective protection measures are essential. Today more than 100 000 sites are protected, cover-ing over 12.7% of the world’s terrestrial area according to the World Database on ProtectedAreas (WDPA). The size of these areas provides, however, little information on how effectivethey are for biodiversity conservation [2]. For example, while in Europe only very small sur-faces of intact old-growth forests (0.2% of the total forest area) are still present [3], in Romaniathe loss of old-growth forest area is an on-going process. Moreover, 72% of the exploitations ofold-growth forests are taking place in protected areas [4].

Lichens are well known bio-indicators and have been frequently used to infer habitat conti-nuity [5, 6]. Studies carried out in habitats with increasing anthropogenic pressure show thatlichens have more differentiated patterns of species diversity related to habitat change thanother groups of organisms in the same sites [7, 8]. Lichen communities differ greatly in naturaland secondary forests in terms of both species richness and composition [9]. Among lichens,the red-listed species are the most threatened by habitat management, and their decline oftenindicates substantial changes in lichen diversity and composition generally [10].

Functional traits have rarely been used to characterize the responses of lichens to habitatmanagement or other environmental variables. However, the species richness of the functionalgroups, i.e. their rarity and substrate specialization in relation to total species richness, havebeen analysed and new patterns determined [11, 12]. The differences found between the lichencommunities in conserved and managed habitats are often connected to substrate preferencesand the distribution area of the species [12, 13]. By comparing the responses of lichen commu-nities from different substrate types, complex diversity patterns could be derived. The mainsubstrate types include living trees, deadwood, soil and rocks, which host different lichen com-munities [14]. Nevertheless, few studies of lichen diversity have yet considered multiple sub-strate types.

The study area, which is part of Rodnei Mountain National Park, shelters a high biodiver-sity, harbouring the highest number of endemic vascular plant species in the South-EasternCarpathians [15] and numerous glacial relicts and rare species of flora and fauna. It is situatedin the northern part of the Eastern Carpathians (Fig 1) and is one of the three UNESCO Bio-sphere Reserves in Romania, included in the Natura 2000 networks of the European Union[16]. A detailed description of the lichen flora of Rodnei Mountains shows that 442 lichen taxahave been reported in the Rodnei Mountains region [17, 18], including a high number ofthreatened lichens.

The aim of this study was to analyse how vegetation type, lichen substrates and environ-mental parameters influence lichen species richness, community composition and lichen eco-logical traits in the protected area of the Rodnei Mountains National Park. We also studiedhow management affects lichen community structure. Current land-use in the Rodnei Moun-tain National Park is either part of the traditional land-use system, e.g. transhumance systemswith seasonal cattle grazing in alpine and subalpine meadows, or of the largely uncontrolledtimber exploitation in protected areas.

Materials and Methods

Study areaThe altitude of the National Park varies between 500–2.303 m a.s.l. The moderately continentaland slightly North Atlantic climate of the region is influenced by the East-West orientation of

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 2 / 16

Competing Interests: The authors have declaredthat no competing interests exist.

the mountain ridges [19]. This mainly impacts the north- and south-exposed slopes, which dif-fer in temperature and precipitation regimes. The mean annual temperatures vary from 6°C atthe base of the mountains to -1.5°C on the ridges at around 2300 m a.s.l. Average annual rain-fall ranges from 1200 mm in the lower regions to over 1400 mm at higher sites on the moun-tain sides [20]. The geology mainly consists of crystalline schist substrata [16].

Management history varies among vegetation types. In the alpine meadows it is generallytraditional sheep grazing, but at lower altitudes also includes horses and cattle. In the foreststhe management history is more complex. Until 1948, the management varied according to theowners’ interests and sometimes no management plan was followed. At the beginning of thetwentieth century and between the twoWorld Wars, forest exploitation involved mainly clearcutting large areas with no special care taken to ensure forest regeneration. After the SecondWorld War, management planning focused on clear cutting small plots, and replanting gapswith spruce, so that the mixed forests began to lose ground. After 1990, illegal cutting becamewidespread and was difficult to stop due to the economic and social situation in the area [16].Given the current largely unregulated management within the borders of the National Park,old-growth forest stands are being transformed into clear-cut areas or managed forests withless structural diversity and habitat heterogeneity, including fewer old and mature trees [4].

Fig 1. Location of the Rodnei Mountains National Park Romania, with the national park boundary, together with the sampling plots, represented bydifferent symbols, according to the different categories of vegetation type andmanagement.

doi:10.1371/journal.pone.0145808.g001

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 3 / 16

Sampling methodIn our study we investigated three vegetation types typical of the National Park, summarised inTable 1, distinguishing two classes of management namely conserved or managed, for each. Ourfield campaign was authorized by the Rodnei Mts. National Park Administration (entry permitsigned by Prof. Gheorghe Coldea, Scientific Director of the National Park) with permission tohave access and collect data (including biological material). No permission was required for col-lecting sites outside the Park borders as they belong to the public domain and are freely accessible.

The samples were collected using a replicated design, consisting of seven circular plots of onehectare in size or 56.4 m in radius for each habitat type and corresponding levels of conservation,i.e. 42 plots in total. The minimum distance between the plots was 100 m. In each sampling plotwe registered the altitude and the geographic coordinates. The aspect and slope were inferred inGIS from a 30 m cell raster, available from http://www.jspacesystems.or.jp/en_/. Aspect was arc-sine transformed, and the exposition expressed as ‘Eastness’ and ‘Northness’ [21].

Six collecting sites in each sampling plot served as the starting point (Table 2) for four lichenrelevés (each with a total area of 50x40 cm) according to the method described in Scheideggeret al. [22]. One relevé was conducted on each of the following substrates: rocks, soil, bark of liv-ing trees and deadwood. All lichens were considered apart from the crustose lichens on therocks. In the plots where some substrate types were not available or were not colonized bylichens, substitute relevés were made on other substrates hosting lichens, resulting in 24 relevésfor each plot.

Data AnalysisWe analysed lichens from all substrates taken together, as well as separately from soil, tree anddeadwood. Since lichens from rocks were often poorly developed in the forest, we only assessedthe saxicolous macrolichens, but did not analyse this group of species separately. Lichen speciesrichness and composition were analysed from relevés on the one-ha plots pooled together.

The variation in lichen species richness, specialist species and red-listed species richnesswere analysed with Generalized Linear Models (GLMs) using Poisson distribution. The depen-dent variable was the number of species belonging to the group of interest. Poisson regressionprovides a model that describes how the mean response of species richness changes as a

Table 1. Description of environmental variables used in the lichen biodiversity analyses.

Measuredvariables

Description of categorical variables No. of samples

Management type Conserved sites (no human exploitation) 21

Managed sites (pasturing in the meadows and logging in theforests)

21

Vegetation type Alpine vegetation (consisting of alpine meadows with barerocks)

14

Spruce forests 14

Mixed forests (composed of beech, spruce, maple, and fir) 14

Description of continuous variables Range (min-max)

Altitude Elevation a.s.l. (in meters) 863–2193

Slope Mean inclination of the slope (in degrees) 4.958–43.846

Aspect Northness -0.99779–0.977885

Eastness -0.99982–0.997849

doi:10.1371/journal.pone.0145808.t001

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 4 / 16

function of one or more explanatory variables. The “log” link function was used. A previousmodel selection based on the smallest AICc values using dredge function from “MuMIn” pack-age [23] was applied to obtain the best model. The predictor variables were checked for collin-earity with the vif function from “Car” package [24]. In order to test the model, chi-square“Lack of fit” test and the dispersion were checked for each GLM. As the terricolous species rich-ness GLM showed over dispersion, we excluded them from the results. The deviance for theGLM fits was analysed with anova.glm function. The p values were adjusted with the “Bonfer-roni”method in all analyses. In the models where categorical independent variables were sig-nificant, we calculated the mean and the standard deviation to see which of the categories hadhigher values. If more than two categories were compared, the Tukey HSD test was used.

Ecological functional traits included substrate preference as defined by Stofer et al. [12]. Weconsidered only the specialist lichens, which are restricted to one substratum. The threatenedred-listed species richness included the following IUCN categories: regionally extinct (RE),critically endangered (CR), rare (R), endangered (EN) and vulnerable (VU). Since no officialRomanian national lichen Red List exists [17], we compiled a list using data from the literature,based mainly on the proposed Red List of macrolichens from Romania (Bartok & Crișan, per-sonal communication) and the available Red Lists of the surrounding Carpathian countries:Ukraine [25], Slovakia [26] and Poland [27].

Variation in lichen species composition was assessed using Partial Canonical Correspon-dence Analysis (pCCA) with the chi-square distance. Vegetation type was chosen as a co-vari-able and all the other measured variables as explanatory variables. The ordination wascalculated using the pCCA function of the “vegan” package [28]. The model selection for theordination was made with a stepwise procedure based on the p values, using the ordistep func-tion from the “vegan” package. The significance of the explanatory variables was analysed withthe ANOVAlike permutation test for Canonical Correspondence Analysis. In order to identifythe characteristic species of the two management categories, a subsequent analysis of the spe-cies composition was carried out with pCCA using only the management type as the explana-tory variable and the vegetation type as the co-variable. Extracting the lichen species scores,those species that were positioned at the ends on the ordination axis were chosen as character-istic for the two types of management: conserved and managed. Only species that occurred inat least four plots were considered in the composition analysis. Statistical analyses were carriedout with R version 3.0.2 (2013 The R Foundation for Statistical Computing).

Results

Species richnessWe found a total of 240 lichen species and one form. Of these, 86 species were substrate spe-cialists (growing on one single substrate) according to the literature [12], 123 species

Table 2. Number, azimuth and distance of the six collecting sites in each one-ha sampling plot asmeasured from the centre of the sampling plot.

Collecting site no. Azimuth (360° gradation) Distance from centre (m)

1 0 18.2

2 60 36.4

3 120 18.2

4 180 36.4

5 240 18.2

6 300 36.4

doi:10.1371/journal.pone.0145808.t002

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 5 / 16

intermediate (growing on two or three different substrates) and 31 species generalists (growingon more than three substrate types). In terms of substrate, we found 102 species on soil, 132species on trees, 93 species on deadwood and 31 species on rocks. The GLMmodel of all lichenspecies showed that the management and vegetation type significantly influenced species rich-ness. Conserved sites had higher mean species richness than managed sites (Table 3). On aver-age, hectare plots in conserved sites harboured 13 species more than managed sites.Comparing the species richness of the vegetation types, mixed forests had the highest speciesrichness per plot, with a mean of 35.6, and the lowest richness was found in the spruce forests,with a mean of 23.5 species per plot.

The findings for lichen species richness on trees were comparable with those for total lichenspecies richness, but the effect of the management type was even stronger. The species richnessin managed forest plots was only 50% of that in conserved plots (Table 3), and that in mixedforests was more than twice as high as that in spruce forests (Table 3). For lichen species ondeadwood, only the vegetation type had a significant effect on species richness, with, asexpected, higher species richness in mixed forests (Table 3).

Species compositionConsidering all substrate types together, 97 species were found on at least four plots and weretherefore used in the composition analysis. Considering the substrate types separately, 40 spe-cies that fulfilled the condition of a minimum of four occurrences at the plot level were foundon soil, 38 on trees, and 24 on deadwood.

The full model of CCA ordination, using all the independent variables measured as explana-tory variables, showed that vegetation type was a strong factor in differentiating lichen commu-nity composition in all our analyses. To determine the effects of the other environmentalvariables measured, we chose the pCCA ordination, using vegetation type as a co-variable(Table 4). The lichen species composition from all substrates taken together was influenced byhabitat management and Northness.

Management type also significantly influenced lichen communities growing on soil, whereasthe other explanatory variables did not. The most characteristic species for conserved commu-nities included Alectoria ochroleuca, Arthrorhaphis citrinella, Cetraria ericetorum, Cladoniagracilis, Cladonia subcervicornis and Lepraria nivalis. In the managed sites, the most character-istic species for the lichen communities from soil were: Cladonia bellidiflora, Cladonia cornuta,Cladonia maxima, Cladonia uncialis and Placynthiella icmalea.

The composition of lichen species from trees was significantly influenced by managementtype, altitude and Northness. The community differentiation of lichens from trees between con-served and managed sites was greater than it was for all lichens, based on the F values of theANOVAlike permutation test (Table 4). The characteristic species of lichen communities fromtrees negatively affected by management included: Chaenotheca chrysocephala, Evernia prunas-tri, Graphis scripta, Lecanora argentata, Lecanora intumescens, Parmeliopsis ambigua, Pertusarialeioplaca and Usnea filipendula. In the managed sites, the representative species of the lichencommunities were:Dimerella pineti, Graphis pulverulenta, Lecanora strobilina,Micarea prasina,Mycobilimbia epixanthoides, Scoliciosporum chlorococcum and Scoliciosporum sarothamni. Thelichen species composition from deadwood substrate varied only with altitude (Table 4).

Substrate specialist lichensOverall we found 86 substrate specialist lichen species (listed in S1 Appendix), namely 47 epi-phytic, 5 lignicolous, 22 terricolous and 12 saxicolous. The GLMs of the substrate specialistlichens were calculated only for all lichens, terricolous lichens and epiphytic lichens.

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 6 / 16

The species richness of the substrate specialist lichens was influenced by the managementtype, vegetation type and their interaction. The conserved habitats hosted a larger number ofsubstrate specialist lichens. Regarding the vegetation type, the species richness in spruce forestswas significantly lower than in mixed forests and alpine vegetation (Table 5). Terricolous sub-strate specialists varied significantly in species richness in the three vegetation types and in theinteraction of management with vegetation type. The alpine vegetation had a mean of 7.07 spe-cies, which is far more than the two forest types, with 0.42 species in the spruce forests and 0.07in the mixed forests. The species richness of epiphytic lichens was influenced by both manage-ment, with more species in the conserved sites, and by vegetation type, with more species in themixed forests (Table 5).

Table 3. Generalized linear models (GLMs) of lichen species richness in relation to the environmental variablesmeasured. The significant variablesare in bold. The mean number and the standard deviation (SD) of species richness per plot for each category of significant factor are displayed in the two col-umns on the right. When all lichens were assessed, the Tukey HSD test showed significant differences according to vegetation type only between the speciesrichness of mixed and spruce forests.

Species richness Model adjR^2 Dispersion Df Deviance Pr(>Chi) Species richness Mean ± (SD)

All lichens 0.922 1.068 Conserved sites 35.8 ± (7.2)

Management type 1 64.623 <0.001. Managed sites 22.5 ± (7.4)

Vegetation type 2 35.685 <0.001. Alpine vegetation 28.28 ± (7.22)

Management type: Vegetation type 2 7.147 0.084 Mixed forests 35.6 ± (8.71)

Spruce forests 23.5 ± (9.9)

Lichens on trees 0.98 1.02 Conserved sites 21 ± (9.51)

Management type 1 48.346 <0.001. Managed sites 10.6 ± (5.1)

Vegetation type 1 61.408 <0.001. Mixed forests 21.8 ± (9.07)

Spruce forests 10 ± (4.22)

Lichens on deadwood 0.393 0.836 Mixed forests 13.1 ± (3.63)

Altitude 1 0.0716 1.000 Spruce forests 10.2 ± (3.38)

Eastness 1 2.2489 0.535

Slope 1 2.4814 0.461

Vegetation type 1 9.1276 0.010

doi:10.1371/journal.pone.0145808.t003

Table 4. Variation in lichen species composition in relation to the environmental variables, measured with ANOVAlike permutation test.

Data Environmental variables DF F N. Perm Pr (>F)

All lichens Northness 1 1.6125 99 0.03

Management type 1 2.4765 99 0.01

Condition (Vegetation type) 2

Lichens on soil Management type 1 2.4033 99 0.01

Condition (Vegetation type) 2

Lichens on trees Slope 1 1.4837 999 0.085

Altitude 1 1.6567 199 0.04

Northness 1 2.068 99 0.01

Management type 1 2.7487 99 0.01

Condition (Vegetation type) 1

Lichens on deadwood Northness 1 1.8623 399 0.0575

Altitude 1 1.8964 99 0.03

Condition (Vegetation type) 1

doi:10.1371/journal.pone.0145808.t004

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 7 / 16

Red-listed speciesOur species list included 99 red-listed lichens (given in S2 Appendix). These are mainly rarespecies, of which 30 were found growing on soil, 58 on trees and 27 on deadwood.

When all substrates were analysed together, the species richness of the red-listed lichens wassignificantly lower in managed plots, with a 50% reduction in managed compared to conservedplots (Table 6). A comparison of the three vegetation types revealed that mixed forests had asignificantly higher number of red-listed species than spruce forests. The highest number ofred-listed species was found in the conserved mixed forests, with a mean of 16.9 species, while

Table 5. Generalized linear models (GLMs) of substrate specialist lichen species richness in relation to the environmental variables measured.The significant variables are in bold. The mean number and the standard deviation (SD) of specialist lichen species richness per plot, for each category of sig-nificant factors, are displayed in the two columns on the right. When all lichens were assessed according to vegetation type, the Tukey HSD test showed sig-nificant differences between the specialist richness of lichen species in spruce forests and that of both alpine vegetation and mixed forests. For terricolouslichens, the specialist lichen species richness of alpine vegetation was significantly different from that of both spruce forests and mixed forests.

Specialist lichens sp. richness model adjR^2 Dispersion Df Deviance Pr(>Chi) Species richness Mean ± (SD)

All lichens 0.95 0.76 Conserved sites 9.52 ± (4.19)

Management type 1 28.62 <0.01. Managed sites 5.09 ± (4.42)

Vegetation type 2 80.88 <0.01. Alpine vegetation 11.14 ± (2.87)

Management type: Vegetation type 2 17.70 0.014 Spruce forests 2.64 ± (2.53)

Mixed forests 8.14 ± (4.27)

Terricolous lichens 0.98 0.6 Alpine vegetation 7.07 ± (2)

Management type 1 0.945 0.99 Spruce forests 0.42± (0.85)

Vegetation type 2 1.755.592 <0.01. Mixed forests 0.07 ± (0.26)

Management type: Vegetation type 2 9.012 0.033

Epiphytic lichens 0.95 0.57 Conserved sites 6.14 ± (4.67)

Management type 1 27.409 <0.01. Managed sites 2.35 ± (2.2)

Vegetation type 1 60.411 <0.01. Spruce forests 1.42± (1.22)

Northness 1 0.598 1.000 Mixed forests 7.07 ± (3.97)

doi:10.1371/journal.pone.0145808.t005

Table 6. Generalized linear models (GLMs) of red-listed (RL) lichen species richness in relation to the environmental variables measured. The sig-nificant variables are in bold. The mean number and the standard deviation (SD) of RL lichen species richness per plot, for each category of significant fac-tors, are displayed in the two columns on the right. In the case of vegetation type when all lichens were assessed, the Tukey HSD test showed significantdifferences only between the RL species richness of mixed and spruce forests. For terricolous lichens, the RL lichen species richness of alpine vegetation dif-fered significantly from that of both spruce and mixed forests.

RL sp. richness model adjR^2 Dispersion Df Deviance Pr(>Chi) Species richness Mean ± (SD)

All lichens 0.92 1.00 Conserved sites 11.47 ± (5.7)

Management type 1 61.163 <0.01. Managed sites 4.71 ± (2.57)

Vegetation type 2 40.152 <0.01. Alpine vegetation 8.85 ± (3.37)

Management type: Vegetation type 2 8.544 0.014 Spruce forests 4.5 ± (3.46)

Mixed forests 10.93 ± (7.14)

Terricolous lichens Alpine vegetation 7.21 ± (2.72)

Vegetation type 0.961 1.544 2 122.724 <0.01. Spruce forests 0.64 ± (1.15)

Management type 1 4.675 0.092 Mixed forests 1 ± (1.47)

Management type: Vegetation type 2 7.08 0.087

Epiphytic lichens 0.985 1.02 Conserved sites 8.14 ± (6.3)

Vegetation type 1 50.238 <0.01. Managed sites 1.57 ± (1.01)

Management type 1 68.156 <0.01. Spruce forests 2 ± (1.96)

Mixed forests 7.71 ± (6.5)

doi:10.1371/journal.pone.0145808.t006

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 8 / 16

the lowest number was found in managed spruce forests, with a mean of 2.1 species (data notshown). Terricolous red-listed lichen richness was influenced only by vegetation type. Thenumber of red-listed species in alpine vegetation was significantly higher (7.2) than in themixed and spruce forests (1 and 0.6, respectively). Red-listed epiphytic lichen richness wasinfluenced by forest management and vegetation type, with a strong discriminating response.Conserved sites harboured over 5 times more red-listed species than managed sites (Table 6).The difference between the two forest types is also high, as can be seen in Table 6.

Discussion

Lichen species richness and compositionAlthough management favours some species, this gain by no means compensates for the muchhigher loss it causes. Previous studies have also reported a decrease in lichen species richnessrelated to management intensity [12, 13, 29, 30].

The mixed forests in our study area contain the greatest number of lichen species, and thecurrent management system with selective cutting is a major source of diversity loss in theseforests. Selective cutting decreases forest structural diversity [31, 32] which is an important fac-tor for lichens, especially epiphytic ones, through various mechanisms. Studies on epiphyticspecies richness from primeval forest [33] or old growth forests [34] have shown that foreststructural diversity with considerable variation in the age of trees and degree of canopy closureis important for the richness of these lichen species. Epiphytic lichens select their substrateaccording to the bark properties i.e. corrugation, pH, moisture-holding capacity and nutrientstatus of the bark [35, 36], which varies with the age of the trees. The distribution of forestlichen species is also influenced by the light regime [37], which relates to canopy closure.

Harvesting mature and old trees mainly affects lichen species that depend on keystonestructures [38] found only in old trees, such as bark crevices or soft bark with a high water-stor-age capacity. They may be found in old beech trees or rain-protected, slightly overhangingtrunks of trees with an asymmetric canopy structure [39]. Species within the study area thatdepend on old trees and that appear to be significantly affected by management include: Artho-nia vinosa, Chaenotheca brachypoda, Heterodermia speciosa, Hypogymnia vittata, Lecanoracinereofusca, Lobaria pulmonaria, Loxospora cismonica,Megalospora tuberculosa,Menegazziaterebrata, Thelotrema lepadinum, Usnea florida, and Usnea fulvoreagens [40, 41].

The other forest management type, clear cutting, characteristically in the spruce forestsinvestigated, temporarily removes the habitat of epiphytic lichen substrates even if it is appliedto relatively small patches. During subsequent forest growth, the even-aged stands vary very lit-tle in their bark substrate types, and light availability is limited for several decades. These dras-tic changes in environmental conditions target the epiphytic lichens. Their mean speciesnumber on trees in the one-hectare plots decreased by 48% after clear-cutting started in 1958[16].

A recent study detected no differences in species richness between unmanaged and man-aged sites [42] in formerly managed forests in Germany. The unmanaged forest stands theysurveyed, however, had not yet reached the level of structural diversity characteristic of old-growth forests and thus still showed signs of former management. This situation is rather dif-ferent from that of the forests in our study.

Comparing the response of lichen species richness on trees and deadwood in mixed andspruce-dominated forests, only lichen species from trees are highly sensitive to forest manage-ment in previously unmanaged, natural forests (Table 3). Species richness on deadwood wasnot affected by forest management. Therefore, we presume that stumps are important for spe-cies richness in the managed forests, as has frequently been claimed for lignicolous lichens

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 9 / 16

[43]. Another reason could be that stumps in managed forests can act as short-term refuges forspecies which can grow not only on living trees but also live on deadwood. For example, in Fen-noscandia and the Pacific Northwest of North America, Spribille et al. [44] found that 43% ofthe lichens growing on trees there can also grow on deadwood.

Management had negative effects not only on species richness but also on the communitycomposition, which may be significantly changed by management.

The lichen communities on trees proved to be very sensitive to the changes brought aboutby forest management. They confirmed the high indicator properties of communities in rela-tion to the intensity of the management described in previous studies [13, 45, 46]. The environ-mental variables represented by altitude and Northness, which are substitutes for climaticconditions in terms of temperature gradients [47], precipitation [20] and light availability, alsocontributed to variations in the community composition. When these environmental variableswere exclude from our constrained ordination (pCCA), the characteristic species of the twomanagement types helped us visualise important processes taking place in the forests studied.The lichen communities from conserved sites mainly consist of common forest species. A highlevel of richness of Caliciaceae family is considered to indicate ancient forests [36], but onlyChaenotheca chrysocephala was frequent in our conserved forests, and this species is rathercommon.

The frequency of some species, such as Usnea filipendula, which are commonly found acrossEurope, declines with forest exploitation both in Europe generally [41] and in our study area.Our conserved forests also contain rare or overlooked characteristic species, such as Lecanoraargentata and L. intumescens [48]. It is, however, surprising that the indicator species charac-teristic of old-growth forests are not represented better in our lichen community assessment inthe conserved forest, given the high number of such species found in our research area [17]. Inthe community composition analysis, we filtered out those species that occurred in less thanfour plots and excluded them from our assessment. The analysis showed that the lichen speciesthat are most characteristic of old-growth forest were not very frequent in our study area. Weassume the reason for this lack of frequency is probably that the managed forests fragment theconserved forests so that the dispersal of old-growth forest species is limited by ecologicalbarriers.

Characteristic lichen communities on trees in managed forests are dominated by commonearly successional lichen species, which are widely distributed, [41]. The ecological preferencesof the lichen species Dimerella pineti,Micarea prasina and Scoliciosporum chlorococcum showthat the managed forests we investigated are predominantly shaded. The limited light repre-sents a filter for a large number of lichen species.

The composition variability of lichens growing on soil requires careful interpretationbecause the alpine vegetation category is not properly distributed across altitude. The con-served sites are at higher altitudes than the managed sites (S3 Appendix), which could inducevariability in the lichen composition. Sheep grazing may affect the composition of the lichencommunity positively or negatively, as some new species may be added while other more sensi-tive species are eliminated through trampling [49]. The trampling effect is even stronger withgrazing cattle [50], as is the case in our sample plots at lower altitudes. Moreover, plantsrespond to grazing in different ways: with some species the abundance increases because theyhave developed a resistance to this type of disturbance [51]. Thus the competition with theplants is stronger in the managed alpine meadows for the terricolous lichens. The characteristiclichens found in the conserved meadows are typical alpine species, some of them rare (e.g. Alec-toria ochroleuca, Cetraria ericetorum, and Cladonia subcervicornis). Their growth could berestricted by the stronger competition from the vascular plants at lower altitudes [52].

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 10 / 16

The managed alpine site communities include characteristic lichen species with wider distri-bution ranges (i.e. Cladonia bellidiflora and C. cornuta), or ruderal species such as Placynthiellaicmalea, which has adapted to a wide range of substrates and is known as a primary coloniser[48].

The lichen communities from soil substrates in the forest did not appear to differ in themanaged and unmanaged plots, but unlike in alpine vegetation, soil is not among the mainsubstrates for lichens in forests due to their low competition abilities [53].

Management affects mainly red-listed and substrate specialist lichenspeciesOur results showed that management reduced the mean level of threatened species richness by59%, taking all lichen species into consideration, and by 81% if only epiphytic lichens were con-sidered. This suggests that, among all lichen species, threatened lichens are the most affectedby forest exploitation, with all the subsequent changes it brings about. This is in accordancewith previous findings [10, 54, 55]. The substantial decline in threatened lichen species takingplace in our study area is alarming considering that this was observed in an overall protectedarea, where any tree cutting is largely an unregulated and unplanned action. The study regionwas declared a biosphere reserve in 1979 and was progressively enlarged up to 2003, to main-tain the regional biodiversity [16] in an otherwise intensively managed forest landscape. Thisdecline is the result of past forest management that spanned several decades, but also the onepracticed nowadays.

Some specialised lichens have strong substrate preferences and grow only on one substratetype. At the opposite extreme are the generalist lichens, which grow on at least three types ofsubstrates [12]. In a biodiversity assessment study, Stofer et al. [12] used functional traits oflichens to describe the significant increase in species richness of generalist lichens along ananthropization gradient. Unlike in our study, they found no clear distribution pattern for thespecialised lichens in the major ecoregions of continental Europe. In our case, the responses ofsubstrate-specialist lichens indicated that the number of all lichens, as well as of terricolous andepiphytic lichens, was significantly smaller in managed sites than in conserved sites. Thisclearly implies that habitat management affects the regional species pool of lichens.

We found several specialised epiphytic lichens on deadwood substrate in the conserved for-ests, namely Candelariella reflexa, Graphis pulverulenta, Lecidella subviridis, Opegrapha viridis,Parmelia submontana and Porina aenea. This suggests that there is continuity between livingtree and deadwood substrates, and the deadwood decadal stages in these forests are verycomplex.

Conservation of lichen diversityAlthough we found that management decreased overall lichen diversity, effective conservationmeasures differ between forest and grassland ecosystems. For terricolous lichens from thealpine meadows grazing must be controlled in intensively grazed pastures to allow the typicallichen communities to persist [56]. In some habitats, grazing is important to sustain higherlichen diversity and save some species from local extinction [57]. In our study region, the diver-sity of terricolous lichen was changed, even by low intensity grazing. In order to conserve thetypical alpine lichen flora, only grazing by wild animals should be allowed, and pasturing bydomestic animals should be restricted. This is the case in the strictly protected alpine meadowsin the Rodnei Mountains National Park, where the lichen diversity is high thanks toconservation.

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 11 / 16

Conservation measures for forest epiphytic and lignicolous lichens depend on the forestmanagement type [58]. For managed forests, the most important recommendation is to extendthe rotation periods of forest exploitations extended [59–61] and of selective cutting in order tocreate structurally heterogeneous landscapes with a range of different habitats [62]. However,in most European forest types, lichen diversity, especially that of threatened and red-listed spe-cies, depends on the presence of old trees [34, 42, 55, 63], because they provide keystone struc-tures for up to 70% of rare and threatened lichen species [39, 40, 64]. Old-growth foreststherefore play an important role as lichen sanctuaries and refuges because they maintain viablelichen populations and serve as sources of lichen propagules for neighbouring managed habi-tats [65]. Old-growth forests thus merit a high conservation priority and should be strictlyprotected.

Habitat size and heterogeneity influence lichen diversity [10, 66]. Since species drift is likelyto increase in small and isolated fragments of habitat due to the dispersal limitation of the spe-cies set [67], the primary focus of conservation strategies for lichens should be, according toScheidegger andWerth [65], the maintenance of habitat quality, connectivity and size. Thisfocus fits in well with the objectives of the protected area in the National Park, which is also aUNESCO Biosphere reserve.

At present, the strictly protected areas (category I IUCN) in the National Park includemainly alpine meadows, shrubs and spruce forest vegetation belts, while mixed forests are notwell represented [16]. The vegetation is distributed in 300–500 m belts along the altitudinalgradient, but mixed forests are only present in the altitudinal interval of 650–1100 m [19].Moreover, the zonation of protection levels in the National Park includes peripheral areaswhere forest exploitation is allowed, which often affects mixed forests as they are mostly con-fined to lower altitudes.

ConclusionsLichen diversity is greater in conserved than in managed sites. Management decisions are thusvery important as they influence the human impact on the vegetation types (i.e., alpine mead-ows and the forests) of this UNESCO Biosphere Reserve. Careful decisions in the managementof natural resources should be taken not only within the borders of protected areas, but overall,in order to diminish the diversity loss as much as possible.

The protected area investigated includes forests that, although under protection status inpresent, have a history with intensive exploitation [16], and have low lichen diversity. Con-versely, some of the currently well-preserved mixed forests with high lichen diversity are out-side its borders (Fig 1) and urgently require protection status. We therefore stronglyrecommend maintaining the high level of protection within the current boundaries of the pro-tected areas and re-evaluating the boundaries of the national park. The re-evaluation is neces-sary especially for the category I IUCN protection zone, to ensure a sufficiently large area ofrepresentative vegetation types of the National Park.

Supporting InformationS1 Appendix. Specialist lichen species ordered according to substrate type.(DOCX)

S2 Appendix. Threatened red-listed lichens considered in our analysis, with the proposedRed List of macrolichens from Romania (RL Ro), the Red List of Ukraine (RL Uk), Poland(RL Pl) and Slovakia (RL Sl). The following IUCN categories are abbreviated in the table:regionally extinct (RE), critically endangered (CR), (R) rare, endangered (EN), vulnerable

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 12 / 16

(VU).(DOCX)

S3 Appendix. Mean values and standard deviation (SD) of the continuous independent var-iables for each category of sampled site.(DOCX)

AcknowledgmentsWe are grateful to Gyongyi Ardelean, Gavril Ardelean, Tudor Mardan, Andrei Meleg, MihaiMiclăuș and Adriana Olenici for their fieldwork assistance, and all others who facilitated thedata collection procces in the field. We thank Silvia Dingwall (Birmensdorf, Switzerland) forediting the manuscript.

Author ContributionsConceived and designed the experiments: IVA CS. Performed the experiments: IVA CK CS.Analyzed the data: IVA CS. Contributed reagents/materials/analysis tools: IVA CK CS. Wrotethe paper: IVA CS.

References1. Barnosky AD, Matzke N, Tomiya S, Wogan GUO, Swartz B, Quental TB, et al. Has the Earth’s sixth

mass extinction already arrived? Nature. 2011; 471:51–7. doi: 10.1038/nature09678 PMID: 21368823

2. Chape S, Harrison J, Spalding M, Lysenko I. Measuring the extent and effectiveness of protected areasas an indicator for meeting global biodiversity targets. Philosophical Transactions of the Royal SocietyB: Biological Sciences. 2005; 360:443–55.

3. Frank D, Finckh M, Wirth C. Impacts of land use on habitat functions of old-growth forests and their bio-diversity. In: Wirth C, Gleixner G, Heimann M, editors. Old-Growth Forests Function Fate and Values.Berlin and Heidelberg, Germany: Springer; 2009. p. 429–50.

4. Knorn J, Kuemmerle T, Radeloff VC, KeetonWS, Gancz V, Biriș I-A, et al. Continued loss of temperateold-growth forests in the Romanian Carpathians despite an increasing protected area network. Environ-mental Conservation. 2012; 40(2):182–93.

5. Frey E. Die anthropogenen Einflüsse auf die Flechtenflora und vegetation in verschiedenen Gebietender Schweiz. Ein Beitrag zum Problem der Ausbreitung undWanderung der Flechten. Veröffentlichun-gen des Geobotanischen Institutes Rübel, ETH Zürich. 1958; 33:91–107.

6. Rose F. Lichenological indicators of age and environmental continuity in woodlands. In: Brown DH,Hawksworth DL, Bailey RH, editors. Lichenology: progress and problems. London: Academic Press;1976. p. 279–307.

7. Jonsson BG, Jonsell M. Exploring potential biodiversity indicators in boreal forests. Biodiversity andConservation. 1999; 8:1417–33.

8. Watt AD, Bradshaw RHW, Young J, Alard D, Bolger T, Chamberlain D, et al. Trends in biodiversity inEurope and the impact of land use change. In: Hester R, Harrison RM, editors. Biodiversity underThreat Cambridge: Royal Society of Chemistry; 2007. p. 135–60.

9. Bergamini A, Scheidegger C, Stofer S, Carvalho P, Davey S, Dietrich M, et al. Performance of macroli-chens and lichen genera as indicators of lichen species richness and composition. Conservation Biol-ogy. 2005; 19:1051–62.

10. Scheidegger C, Goward T. Monitoring Lichens for Conservation: Red Lists and Conservation ActionPlans. In: Nimis PL, Scheidegger C, Wolseley P, editors. Monitoring with Lichens—Monitoring Lichens.Dordrecht, Boston, London: Kluwer Academic; 2002. p. 163–81.

11. Pinho P, Bergamini A, Carvalho P, Branquinho C, Stofer S, Scheidegger C, et al. Lichen functionalgroups as ecological indicators of the effects of land-use in Mediterranean ecosystems. Ecological Indi-cators. 2012; 15:36–42.

12. Stofer S, Bergamini A, Aragon G, Carvalho P, Coppins BJ, Davey S, et al. Species richness of lichenfunctional groups in relation to land use intensity. The Lichenologist. 2006; 38(4):331–53.

13. Nascimbene J, Marini L, Nimis PL. Influence of forest management on epiphytic lichens in a temperatebeech forest of northern Italy. Forest Ecology and Management. 2007; 24:43–7.

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 13 / 16

14. Seaward MRD. Environmental role of lichens. In: Nash THI, editor. Lichen biology. 2 ed. Cambridge:Cambridge University Press; 2008. p. 274–98.

15. Hurdu BI, Puscas M, Turtureanu PD, Niketic M, Coldea G, Zimmermann NE. Patterns of plant ende-mism in the Romanian Carpathians (South-Eastern Carpathians). Contribuţii Botanice. 2012; 47:25–38.

16. APNMR. Management Plan 2013 Rodnei Mts. National Park, Biosphere Reserve (NATURA 2000 SACand SPA) (In Romanian)2013.

17. Ardelean IV, Keller C, Scheidegger C. Lichen flora of Rodnei Mts. National Park (Eastern Carpathians,Romania) including new records for the Romanian mycoflora. Folia CryptogamicaEstonica. 2013;50:101–15.

18. Ciurchea M. Key to Lichens from Romania (In Romanian). Iași: Bit; 2004.

19. Coldea G. Muntii Rodnei: Studiu geobotanic. Bucuresti: Ed. Academiei Romane

20. Gorduza V. Physico–geographical characterisation of the Pietrosu Rodnei Nature Reserve. PietrosulRodnei at 50 years (In Romanian). Cluj-Napoca–Baia Mare: Academia RSR 1983. p. 56–66.

21. Roberts DW. Ordination on the basis of fuzzy set theory. Vegetatio. 1986; 66:123–31.

22. Scheidegger C, Groner U, Keller C, Stofer S. Biodiversity Assessment Tools–Lichens. In: Nimis PL,Scheidegger C, Wolseley PA, editors. Monitoring with Lichens–Monitoring Lichens. Dordrecht, Boston,London: Kluwer Academic; 2002. p. 359–65.

23. Bartoń K. Multi-model inference: CRAN; 2014.

24. Fox J, Weisberg S. An {R} Companion to Applied Regression: CRAN; 2014.

25. Didukh YA. Red data book of Ukraine, Vegetable kingdom (In Ukrainian): Globalconsalting; 2009.

26. Pisút I, Guttová A, Lackovicová A, Lisická E. Cerveny zoznam lisajníkov Slovenska (December 2001)[Red List of lichens of Slovakia (December 2001)] Ochrana Prírody. 2011; 20:23–30.

27. Cieslinski S, K. C, Fabiszewski J. Red List of extinct and threatened lichens in Poland. In: K. C, editor.The threat to lichens in Poland. 91: Monographiae Botanicae; 2003. p. 13–49.

28. Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’Hara RB, et al. Community EcologyPackage: CRAN; 2013. Available from: http://cran.r-project.org, http://vegan.r-forge.r-project.org/.

29. Ellis CJ. Lichen epiphyte diversity: A species, community and trait-based review. Perspectives in PlantEcology, Evolution and Systematics. 2012; 14:131–52.

30. Nascimbene J, Marini L, Ódor P. Drivers of lichen species richness at multiple spatial scales in temper-ate forests. Plant Ecology & Diversity. 2012; 5(3):355–63.

31. Boncina A. Comparison of structure and biodiversity in the Rajhenav virgin forest remnant and man-aged forest in the Dinaric region of Slovenia. Global Ecology and Biogeography. 2000; 9(3):201–11.

32. Humphrey JW, Davey S, Peace AJ, Ferris R, Harding K. Lichens and bryophyte communities of plantedand semi-natural forests in Britain: the influence of site type, stand structure and deadwood. BiologicalConservation. 2002 10//; 107(2):165–80.

33. Dymytrova L, Nadyeina O, Hobi ML, Scheidegger C. Topographic and forest-stand variables determin-ing epiphytic lichen diversity in the primeval beech forest in the Ukrainian Carpathians. Biodiversity andConservation. 2014; 23(6):1367–94.

34. Nascimbene J, Marini L, Motta R, Nimis P. Influence of tree age, tree size and crown structure on lichencommunities in mature Alpine spruce forests. Biodiversity and Conservation. 2009 2009/06/01; 18(6):1509–22. English.

35. Fritz Ö, Brunet J, Caldiz M. Interacting effects of tree characteristics on the occurrence of rare epiphytesin a Swedish beech forest area. The Bryologist. 2009; 112(3):488–505.

36. Selva SB. Lichen diversity and stand continuity in the northern hardwoods and spruce-fir forests ofnorthern New England and western New Brunswick. Bryologist. 1994; 97:424–9.

37. Gauslaa Y, Solhaug KA. Differences in the susceptibility to light stress between epiphytic lichens ofancient and young boreal forest stands. Functional Ecology. 1996; 10(3):344–54.

38. Tews J, Brose U, Grimm V, Tielbörger K, Wichmann MC, Schwager M, et al. Animal species diversitydriven by habitat heterogeneity/diversity: the importance of keystone structures. Journal of Biogeogra-phy. 2004; 31(1):79–92.

39. Scheidegger C, Stofer S. Schlüsselstrukturen, Vernetzung, ökologische Kontinuität. Schweiz Z For-stwes. 2015; 166(2):75–82.

40. Scheidegger C, Stofer S. Flechten imWald: Vielfalt, Monitoring und Erhaltung. Forum für Wissen.2009:39–50.

41. Wirth V. Die Flechten Baden–Württembergs. Stuttgart: Ulmer; 1995.

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 14 / 16

42. Boch S, Prati D, Hessenmöller D, Schulze E-D, Fischer M. Richness of Lichen Species, Especially ofThreatened Ones, Is Promoted by Management Methods Furthering Stand Continuity. PLoS ONE.2013; 8(1):e55461. doi: 10.1371/journal.pone.0055461 PMID: 23383196

43. Caruso A, Rudolphi J, Thor G. Lichen species diversity and substrate amounts in young planted borealforests: A comparison between slash and stumps of Picea abies. Biological Conservation. 2008 1//;141(1):47–55.

44. Spribille T, Thor G, Bunnell FL, Goward T, Björk CR. Lichens on dead wood: species-substrate relation-ships in the epiphytic lichen floras of the Pacific northwest and Fennoscandia. Ecography. 2008;31:741–50.

45. Aragón G, Martínez I, Izquierdo P, Belinchón R, Escudero A. Effects of forest management on epiphyticlichen diversity in Mediterranean forests. Applied Vegetation Science. 2010; 13(2):183–94.

46. Nascimbene J, Marini L, Nimis PL. Epiphytic lichen diversity in old-growth and managed Picea abiesstands in Alpine spruce forests. Forest Ecology and Management. 2010 7/30/; 260(5):603–9.

47. Körner C. The use of ‘altitude’ in ecological research. Trends in Ecology & Evolution. 2007 11//; 22(11):569–74.

48. Smith CW, Aproot A, Coppins BJ, Fletcher A, Gilbert OL, James PW, et al. The Lichens of Great Britainand Ireland. London: British Lichen Society; 2009. 1046 p.

49. Clair LLS, Johansen JR, Clair SBS, Knight KB. The Influence of Grazing and Other Environmental Fac-tors on Lichen Community Structure along an Alpine Tundra Ridge in the Uinta Mountains, Utah, U.S.A. Arctic, Antarctic, and Alpine Research. 2007; 39(4):603–13.

50. Kohler F, Gillet F, Gobat J-M, Buttler A. Effect of cattle activities on gap colonization in mountain pas-tures. Folia Geobot. 2006 2006/09/01; 41(3):289–304. English.

51. Evju M, Austrheim G, Halvorsen R, Mysterud A. Grazing responses in herbs in relation to herbivoreselectivity and plant traits in an alpine ecosystem. Oecologia. 2009 2009/08/01; 161(1):77–85. English.doi: 10.1007/s00442-009-1358-1 PMID: 19412704

52. Bruun HH, Moen J, Virtanen R, Grytnes J-A, Oksanen L, Angerbjörn A. Effects of altitude and topogra-phy on species richness of vascular plants, bryophytes and lichens in alpine communities. Journal ofVegetation Science. 2006; 17(1):37–46.

53. Will-Wolf S, Hawksworth DL, Mccune B, Rosentreter R, Sipman HJM. Lichenized Fungi. In: MuellerGM, Bills GF, Foster MS, editors. Biodiversity of Fungi: Inventory and Monitoring Methods. Burlington,San Diego, London: Elsevies Academic Press; 2004. p. 173–95.

54. Fritz Ö, Gustafsson L, Larsson K. Does forest continuity matter in conservation?–A study of epiphyticlichens and bryophytes in beech forests of southern Sweden. Biological Conservation. 2008 3//; 141(3):655–68.

55. Thor GR. Red-listed lichens in Sweden: habitats, threats, protection, and indicator value in boreal conif-erous forests. Biodiversity & Conservation. 1998 1997/01/01; 7(1):59–72. English.

56. Rai H, Upreti DK, Gupta RK. Diversity and distribution of terricolous lichens as indicator of habitat het-erogeneity and grazing induced trampling in a temperate-alpine shrub and meadow. Biodiversity andConservation. 2012; 21:97–113.

57. Leppik E, Jüriado I, Suija A, Liira J. The conservation of ground layer lichen communities in alvar grass-lands and the relevance of substitution habitats. Biodiversity and Conservation. 2013 2013/03/01; 22(3):591–614. English.

58. Nascimbene J, Thor G, Nimis PL. Effects of forest management on epiphytic lichens in temperatedeciduous forests of Europe–A review. Forest Ecology and Management 2013; 298:27–38.

59. Esseen P-A, Renhorn K-E, Pettersson RB. Epiphytic Lichen Biomass in Managed and Old-GrowthBoreal Forests: Effect of Branch Quality. Ecological Applications. 1996 1996/02/01; 6(1):228–38.

60. Hedenås H, Hedström P. Conservation of epiphytic lichens: Significance of remnant aspen (Populustremula) trees in clear-cuts. Biological Conservation. 2007; 135(3):388–95.

61. Moning C, Müller J. Critical forest age thresholds for the diversity of lichens, molluscs and birds inbeech (Fagus sylvatica L.) dominated forests. Ecological Indicators. 2009; 9:922–32.

62. Rudolphi J, Gustafsson L. Forests Regenerating after Clear-Cutting Function as Habitat for Bryophyteand Lichen Species of Conservation Concern. PLoS ONE. 2001; 6(4):e18639.

63. Scheidegger C, Clerc P. Liste Rouge des especes menacees en Suisse: Lichens epiphytes et terri-coles. Berne, Birmensdorf,: Office fédéral de l’environnement, des forêts et du paysage OFEFP,Berne, Institut fédéral de recherchesWSL, Birmensdorf, et Conservatoire et Jardin botaniques de laVille de Genève; 2002.

64. Caruso A, Rudolphi J. Influence of substrate age and quality on species diversity of lichens and bryo-phytes on stumps. The Bryologist. 2009; 112(3):520–31.

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 15 / 16

65. Scheidegger C, Werth S. Conservation strategies for lichens: insights from population biology. Fungalbiology reviews 2009; 23:55–66.

66. Gignac LD, Dale MRT. Effects of Fragment Size and Habitat Heterogeneity on Cryptogam Diversity inthe Low-boreal Forest of Western Canada. The Bryologist. 2005 2005/03/01; 108(1):50–66.

67. Hubbell SP. The Unified Neutral Theory of Biodiversity and Biogeography. Princeton, Boston: Prince-ton University Press 2001. 375 p.1990.

Management and Lichen Diversity in a Protected Area

PLOS ONE | DOI:10.1371/journal.pone.0145808 December 30, 2015 16 / 16